January 2012 Summaries

The author, Michael, acknowledges the help of many fine people in contributing to his book on creating a full social web application for movie enthusiasts using Spring Data Neo4j. The book covers both a narrative tutorial and a thorough reference to the graph facilities available in Spring Data Neo4j, including core concepts, querying, and simple annotated POJO programming model. It also introduces two alternative facilities: Neo4jTemplate and entity repositories, which offer convenient APIs for working with the Neo4j graph database and extend existing techniques of working with annotations and object mapping.

Rapidus, an online newsletter in Sweden, recently awarded the prize to Neo Technology for their graph-based database. The award recognizes the company's innovative product that has gained international admiration and commercial success. The jury praised Neo Technology's economic platform, which has attracted large investors and new clients, including QlikTech, a well-known company with 22,000 customers worldwide. The CEO of Neo Technology participated in the event via link from Silicon Valley, delivering an entertaining speech that included a joke about persistence. The award was met with joy by the team, who were surprised but thrilled to receive recognition for their work.

The latest release of Neo4j includes a host of great new features, such as the public beta of the Neo4j Add-on for Heroku, improved Cypher syntax and performance, a full Neo4j Shell commands in the web admin interface, and the ability to ensure that key-value pairs for entities are unique. The Lucene upgrade has been upgraded to version 3.5, providing recent enhancements and bug fixes. Breaking changes have been introduced, but older functionality will be supported for two GA releases. A new way of handling breaking changes is also being implemented, flagging them in the change logs as "BREAKING CHANGE." Additionally, a deprecated feature has been removed, and community members have contributed to this release with various development ideas and pull requests. The release includes sharding, supernodes, node types, unique indexing, and N-ary relationships features, which are available now for download.

The Neo4j community is launching a challenge to create the best Heroku-hosted demo or template applications for the Neo4j Add-on. Participants will have the freedom to choose their application type, programming language, frameworks, and Neo4j-driver, with the goal of creating educational, tested, and working well applications. The challenge aims to provide valuable feedback for the Neo4j Add-on while showcasing ready-made applications that can be shared in the Gensen Heroku Template Repository. A guest panel will review top applications to identify winners, who will receive prizes including a Neo4j-Heroku t-shirt.

The difference between RelatedTo and RelatedToVia in Spring Data Neo4j is that RelatedTo refers to the node-entities at the other end of a relationship, while RelatedToVia refers to relationships themselves as relationship-entities. There isn't a built-in facility for "supernode" relationship navigation, but indexing relationship-entities can be used to write an index and retrieve specific data using GraphTraversal and TraversalBuilder. Embedded Neo4j databases cannot be accessed by multiple applications simultaneously due to the limitations of single-JVM access, requiring alternative protocols to be exposed on top of the database for interaction with other processes.

The new Neo4j release improves the API to ensure entity uniqueness for a given key-value pair, making it easier to maintain this uniqueness even when multiple processes are adding data simultaneously. The `putIfAbsent` method and `UniqueFactory` provide thread-safe solutions, allowing users to retrieve an entity using get-or-create semantics. Additionally, array properties can now be queried in Cypher, and aggregation performance has been improved. Neo4j is upgrading its indexing features by moving to a newer stable release of Apache Lucene 3.5, which includes recent enhancements. The release also includes bug fixes and is available as a hosted cloud service or downloadable packages. Developers are encouraged to join the community mailing list to stay up-to-date with new features in the pipeline.

You can deploy a Spring MVC application with Spring Data Neo4j on Heroku by making your application self-hosting, accessing Heroku environment variables for configuration, and deploying to the cloud. To make your application self-hosting, you need to add dependencies for Jetty to your pom.xml file, create a launch script using the appassembler-maven-plugin, and modify the Spring application context to use Neo4j environment variables. After that, you can deploy to Heroku by creating a Procfile, initializing a local git repository, provisioning a Heroku stack, adding the Neo4j Add-on, and deploying your application. Once deployed, you can access your Neo4j graph through Heroku's Webadmin using the NEO4J_URL environment variable.